An adequate estimate of the radiation components on the earth's surface may help reveal many important interactions between the earth's surface and the atmosphere. In-situ measurements of radiation components are sparse, and remote sensing is one way to overcome this limitation. The clouds and the earth's radiant energy system (CERES) provides a long-term estimate of shortwave and longwave radiation over the entire globe. This article compared and evaluated all components of the surface radiation, estimated using CERES SYN1deg Ed3A and SYN1deg Ed4A data (shortwave up and down, longwave up and down, and photosynthetically active radiation) against measurements for 15 sites located in Brazil. Our results indicated that CERES SYN1deg estimates are accurate for all variables evaluated, with the SYN1deg Ed4A version increasing the R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and decreasing the RMSE from the SYN1deg Ed3A version. We also evaluated the main driving factors controlling the variability of the surface radiation components, using cluster analysis and multiple linear regression. The results showed that surface temperature and total precipitable water vapor are the main driving factors affecting the variability of the different radiation components. The results also highlighted the influence of climate conditions and biome features on the estimates of surface radiation components by CERES. The radiation data provided by CERES SYN1deg Ed4A proved to be a promising alternative for large regions where meteorological information is unavailable or sparse.